The exchange of intra- and extra-cellular sodium and potassium ions is essential to cell function and integrity. In the heart, evidence from animal and human radionuclide imaging studies of potassium and potassium analogs, and from biochemical and magnetic resonance spectroscopy (MRS) studies of animal models, indicates that sodium- potassium pump function is compromised during periods of myocardial ischemia and that it is lost in non-viable, infarcted tissue as intra- and extra-cellular pools equilibrate. Sodium (23Na) magnetic resonance imaging (MRI) is uniquely able to image and measure noninvasively naturally abundant, endogenous sodium in the body. 23Na MRI at magnetic fields of > 2.7 Tesla (T) in animal models demonstrate a 2-fold increase in 23Na signal levels in nonviable, histologically-confirmed, acute reperfused myocardial infarction (MI). Owing to its higher tissue concentration and sensitivity and its short relaxation time, 23Na MRI has an enormous sensitivity advantage compared, for example, with the detection of high-energy phosphate metabolites by phosphorus (31P) MRI (approximately 80- fold). Thus 23Na MRI is a potentially unique and important tool for assessing cellular metabolic and ionic function through altered sodium levels in patients with ischemic heart disease and/or MI. Yet 23Na is not now routinely possible on clinical 1.5T MRI scanners. Moreover, human 23Na MRI has never benefitted from new MRI hardware and software technology. In preliminary studies we implemented 23Na MRI on a clinical MRI scanner, and demonstrate altered 23Na MRI levels in MI. We show preliminary stress-23Na MRI data from patients with stress-induced ischemia detected metabolically by 31P MRS. Here we propose to develop and optimize human cardiac 23Na MRI on a clinical 1.5 T MRI/MRS system, by implementing high-speed MRI, 23Na phased-array detection, resolution-enhancement using a priori anatomic information, and methods of suppressing 23Na signals from ventricular blood. We will use optimized 23Na MRI to characterize normal and infarcted human myocardium, and to test the hypotheses that 23Na MRI can differentiate normal from non-viable reperfused MI in patients as detected by radionuclide imaging, and compared with contrast- enhanced MRI. Further, the hypothesis that optimized 23Na MRI can detect stress-induced changes in sodium in energetically- compromised myocardium will be tested in combined stress-23Na/3 1P metabolic studies. The availability of thousands of clinical MRI scanners offers a great opportunity for advancing 23Na MRI as a tool for assessing sodium pump function in human heart disease.